Digital DM — Acceptable Use Policy
Draft — v0.1. Not legal advice.This policy is contractually required. Black Forest Labs' developer terms oblige us to bind end users to an acceptable use policy "no less protective" than their own. It is not optional boilerplate, and it cannot be watered down without breaching our own upstream licence.
The short version
This is a tool for making imaginary people, places, and monsters for a game played with friends. Almost everything you want to do is fine.
The prohibitions below are narrow and they exist for real reasons, not because we are squeamish. Fantasy violence, morally repugnant villains, grim tone, horror, and adult themes in a fiction context are all permitted. We are not going to refuse you an assassin.
Prohibited
You may not use Digital DM, or its output, to:
1. Sexual content involving minors
Any depiction, description, or sexualisation of a minor. This is the one thing with no context that makes it acceptable, no framing that makes it fiction, and no appeal that will change our answer. Accounts are terminated without warning and we report where legally required.2. Sexual or harassing content targeting real people
No sexual, defamatory, or harassing content depicting real, identifiable individuals — public figures included.3. Real-world violence
No content that incites, threatens, plans, or instructs real-world violence against real people or groups. **In-fiction violence against fictional characters is not this**, and we are not going to confuse the two.4. Hate and harassment
No content that promotes hatred against, or dehumanises, real people or groups on the basis of a protected characteristic.5. Deception about AI origin
You may not present generated content as human-authored, or strip the AI-provenance markings from exports and images. **Use it commercially — just don't lie about where it came from.**6. Training competing models
You may not use output to train, fine-tune, or distil an AI model that competes with the models that produced it.7. High-risk use
No use in any context where failure could cause death, injury, or serious property or environmental damage.8. Abuse of the service
No circumventing quotas or rate limits, no unauthorised automation, no scraping or bulk export of the generation database, no reselling access to the underlying models.How this is enforced
Two layers, and they behave differently on purpose:
A narrow, automatic block applies to items 1–3 above, before anything is generated. It is deliberately scoped so that legitimate tabletop content — including dark, violent, or adult fiction — never trips it. If it ever does, that is a bug and we want to hear about it. Everything else is reviewed after the fact. Content is generated, you get it, and a human may later hide it from our public pages. **We do not block your generation because a word looked suspicious.** You are running a game; you should not be arguing with a filter mid-session.Content hidden from public pages remains available to you at its direct link.
Reporting
Report abuse: [CONTACT EMAIL]